© Sadikgulec | Dreamstime.com - US Soldiers In Iraq Photo
In 2001 I wrote an article entitled ‘A Year Observing Command and Control’. It was published in the British Army Review. I had recently spent a year, with what was then the Defence Evaluation and Research Agency, examining command posts (CPs) and how they worked. We looked at CPs in Canada, Corsica, Ramstein and Madrid. We looked at unit, brigade, divisional and theatre-level HQs.
I have been doing the same thing, less formally, ever since. I have followed trends. I have followed individuals (one officer who was a lieutenant colonel when I first watched him is now a four star General). This article reports my main observations from the last ten years or so. That period has seen Western-led coalitions come and go from Iraq and Afghanistan. It has also seen the widespread deployment of digital IT into command posts.
Some years ago someone in the HQ of Multinational Corps, Iraq (HQ MNC-I) made a suggestion for a community engagement project in Baghdad. It took six days just to organise, plan and schedule the briefings required to get a brigadier general to sign it off. On the fifth day someone noticed that one of the insurgent militias was doing exactly what HQ MNC-I was planning to do. The plan had to be scrapped: playing copycat was not a good idea. Someone closely involved said that the HQ was a perfect planning machine, but ‘the wheels never touched the ground’. Plans never became reality. Something was wrong. That observation is typical. There is something badly wrong with our CPs.
Observation: The introduction of IT into CPs has brought new capabilities, but overall, the impact has been negative. HQs have become much bigger and tend to produce worse plans, and take much longer to produce them. Much, but not all of this, is due to digital IT. It has led to a number of problems
Observation: We have been bewitched by the technology. We ascribe things to it which we could do without. For example, a colleague told me that he once got hold of CAS, from another nation, within seven minutes. Wasn’t that amazing? No. In 1944 it was normal to get the fire of up to 12 battalions of field artillery, from 3 different nations, directed onto a target within five minutes. The 21st Army Group routinely did it in Normandy in 1944, with no digital IT. Digital IT does some wonderful things, but some of them can be done with voice radio communications and slick drills and procedures.
Observation: Even before digital IT was introduced, people said that CPs would have to get bigger to support it. That was an error: IT should make working groups more efficient. That should mean doing the same with fewer people. In practice it just meant ‘more people’, which caused problems (see below).
Observation: IT, and particularly word processing packages, enabled people to create more output in terms of the length of documents. As a result the primary output of a CP (the orders it produces) simply got bigger. This is a major problem. Orders became wordy and difficult to understand. Long orders are entirely inconsistent with Mission Command. In practice the meaning of ‘mission command’ has become adulterated.
Observation: The use of digital IT in general, and battle management systems in particular, have fostered a growth in explicit process which has become absurd. Process has become an end in itself. In addition, strict adherence to explicit process is contrary to the aim of helping talented staff officers to work well as a team. People become enslaved to their machines. People can become frustrated. Some indulge in activities that are counterproductive. They also waste a lot of time doing things like answering e-mails. The issue of process is a wider problem, discussed further below.
Observation: CPs now have sufficient bandwidth to be able to pass multiple-page documents around electronically. As described above, products such as orders have become bulkier. Worse, however, is that staff now tend to ‘cut and paste’ information from higher HQ, rather than précis or abstract it before passing it down. In the 1980s many Western HQs did not even have photocopiers, so producing anything more than a few pages long was impractical. In the Falklands conflict all messages had to be passed by Morse code. Both measures meant that the staff had to really think about what they wished to tell subordinates. They had to do it in as few words as possible. As the (Falklands) Force Signal Staff Officer told me, ‘that was a good thing’. That discipline has gone.
Observation: Over the same period, CP’s were freed of the Cold War requirement of having to move frequently. That has meant that the discipline of having to be small has been removed. Naval and amphibious CP’s have had to remain small, and are typically more effective.
Observation: The quality of graphical representation has gone down. That is, not least, because staff officers no longer actually draw their graphics. They are typically created on digital IT (such as Powerpoint). As a result, the staff are not intellectually connected with their product in the same way. NATO’s system of symbology and graphical representation, APP-6, was rewritten to make it compatible with early digital IT (with low resolution or single-colour screens). The result is less intuitive, less elegant and more cluttered. APP-6 should be revised.
Observation: There has been a consistent belief that adding manpower to CPs is a good thing. It is not. It is counterproductive, but that is not obvious. There is an optimum size for groups of human beings who interact. It is a balance between dividing a job up between more people to reduce the time taken, against the increased time needed to brief all the members of a larger group. As CPs get bigger, they get inefficient. They are clearly well beyond their optimum size now. The graph is a representation of Brook’s Law which points to the same result.
Figure 1: Graphical Representation of Brooks’ Law
Observation: Many HQs have more than doubled in size since 2003 or so. There are many apparent drivers for this growth, but most are mistaken. Most HQs do not need genuine 24-hour operation. They need planners who can plan when required. That rarely means as much as 16 hours in 24. Thus, planning in practically any HQ can be done by one man, or at most by a few men working in one shift. Not least, too much nugatory planning is taking place. HQs also need a few ‘current operations’ staff who work around the clock, this requires two shifts.
Comment: If very few insurgents do anything during the hours of darkness, why do COIN HQs all need to be scaled for 100%, 24-hour working? Modern HQs also need individual specialists such as political and media advisors. However, the few such posts actually required does not nearly account for the vast increase in numbers observed.
Observation: The expansion of NATO, and the large number of nations represented in coalitions, has led to a huge increase in the supply of staff numbers. That is faulty rationale to enlarge HQs. Related to that is the growth in staff ranks, discussed below.
Comment: The Danes like to say that, with anything more than 200 people in a CP, it doesn’t need any input. It runs itself! A British officer hearing that observed that all it needs is a process. No, said a Dane, the staff will invent one themselves.
Comment: A senior staff officer was recently overheard to observe that his HQ needed to ‘concentrate on the process, not the structure’. He was partly right, but basically wrong. HQs should concentrate on output, which leads to outcome. Explicit process should be a guide to that. Process has become a bible rather than a guide.
Observation: Most explicit staff processes, such as NATO’s Military Decision Making Process (MDMP), are simply not fit for purpose. They take too long. They produce plans which are mediocre and orders which are too long. An instructor at a Scandinavian staff college recently observed that only the very best staff college students can really use the MDMP properly. That means that something is very, very wrong. It may mean that either most staff college students aren’t good enough; or that the instructors aren’t good enough. Both are unlikely. It actually means that the MDMP is too complex, and therefore not fit for purpose.
Comment: Processes should be streamlined and allowed to evolve, in order to produce good plans and short orders quickly. It is commonly assumed that processes should not be streamlined until the staff have mastered the basics, such as the MDMP as taught. That is a fallacy. Firstly, the evidence is that it will never happen (see above). Secondly, the process just gets in the way. Able and well-intentioned staff can do better without it, if allowed to practice. Just ask yourself: would you know a good, short order describing a good plan if you saw it? If so, just concentrate on producing them. Get rid of all the rest of the junk, such as synchronisation matrices.
Observation: Wargaming Course of Action (COAs) (etc) is an aspect of process which was introduced in the 1990s and 2000s. If the staff have to resort to wargaming COAs, their plans are too complex. The problem may be that they are trying to do things that are too complex and will not survive contact with reality. Alternatively, the staff are not sufficiently well trained. Both issues can, and should be addressed.
Observation: Some senior staff become hyperactive. The G5 of one Coalition HQ in Afghanistan used to work up to 16 hours a day. He produced up to 10 fragmentary orders a day. In practice only about one in 15 affected subordinate units by as much as changing the tasking of a single infantry section. It was a clear case of a triumph of process over outcome.
Comment: The outputs of CPs are plans, represented in orders. Other than the initiating directives for a campaign or major operation, few orders should ever be more than a few pages long in total, perhaps ten or twelve. If they are, they are not fit for purpose. The initiating order for Operation OVERLORD, the Allied landing in Normandy in 1944, was incredibly short: about a dozen pages in total.
Observation: Part of the problem is that of poorly trained staff, both individually and collectively. Many of the staff in a number of major NATO HQs today have not attended a staff college. They shouldn’t be there. The HQ would be better off if they were not there. The small number of trained staff left behind could get on and work together better in smaller, more cohesive groups.
Observation: A few years ago a colleague noticed that the orders originating from Regional Commands (RC’s) in Afghanistan were inconsistent with those originating in HQ ISAF. The ISAF commander had recently produced a revised campaign plan. My colleague visited the RC commanders and gently asked them what they thought of the ISAF campaign plan. Their general opinion was that it was ‘useful’ and ‘insightful’. In simple terms, each RC had written its own campaign plan, largely in isolation.
Comment: The truth is brutally simple. In one theatre there should only be one campaign, and therefore only one campaign plan. Subordinate HQs should extract their orders from that and make tactical plans based on it, especially at the 2-star level. The Operational Level has become badly overdone. All together, developed nations teach hundreds of officers a year how to conduct operational-level planning. An incredibly small number actually need to do it.
Observation: HQs have become so big that it is almost impossible to train them. Staff training has become a big, unwieldy process often conducted only once a year. That is demonstrably insufficient (not least because of the frequent turnover of staff). In the invasion of Iraq in 2003 one British brigade HQ was notably poorly run. It was possibly poorly led, but certainly poorly trained. That was in part because it had been considerably expanded above its peacetime establishment. Unfortunately, the wrong remedy was applied. The HQ didn’t cope well and it was assumed that the reason was that it had been too small. The result was to make all British brigade HQs bigger. It was clearly the wrong answer. The answer should have been better training.
Observation: I know a number of generals personally. I have observed a lot more. Some are good, and definitely do a good job. A few are outstandingly good. Some look, or act, the part. Of those, some are mediocre and some are inept. To be clear, inept commanders have commanded formations on international operations.
Observation: Surprisingly, many generals are not good tacticians. They do not have a clear idea about what to do in order to beat their adversaries. That suggests that they have been poorly trained, or selected for the wrong reasons. Overall, generals are not as good as their nations would expect.
Comment: Generals are almost exclusively promoted in peacetime, using criteria largely based on peacetime requirements. Nowadays many have operational experience, but that is seen as a qualification (a tick in the box). Real operational ability is not in practice a requirement. Much of the reason is due to do with social dynamics which underlie the way that annual appraisals and promotion systems work in practice. It is quite subtle. I know of one General who was groomed for promotion before he even got to staff college, because he looked and acted the part. In my opinion he was overpromoted by four to five ranks.
Comment: My own experience tells me that many of the people promoted to high rank were predicted to do so from staff college, 20 years before. There are people in armies who would have been better, but they are in practice overlooked quite early. That suggests that selection processes are insufficiently perceptive. Many of those who did succeed were moderately competent at junior levels, but looked and acted the part. They were then groomed and gained experience in high-calibre posts. Many nations get this wrong. It would be entirely possible to identify future commanders who could beat others in comparative tests, for example tests based on war gaming.
Quite separately, it is quite clear that almost nobody in any HQ should be above the rank of major. There is virtually no post in an HQ that a major cannot be trained to do: given the right training. There are very, very few posts which require levels of experience which a major cannot be expected to have. Those posts should be held by lieutenant colonels or colonels. There should rarely be more than one or two of those in any HQ; especially the largest. In practice there are often several layers of staff above the level of major. Detailed examination reveals that the officers in those layers are usually counterproductive. They either slow down staff process; overcomplicate it; take decisions that their subordinates know are flawed; tell the commander (and other senior staff officers) things they want to hear, rather than the objective truth; or some combination of all the above. All are detrimental to operational effectiveness.
In addition, the act of inserting more senior staff (for example lieutenant colonels) is demonstrably self-defeating. It means that the majors and captains that they supervise have less responsibility than before. In due course, they will not be as experienced when they become lieutenant colonels themselves. If your majors aren’t good enough, don’t use lieutenant colonels. Select and train your majors better. More people and higher ranks is not the answer. Fewer, but better trained people is the answer.
We should be quite clear. To take one real example, a corps HQ does not need a lieutenant general, two major generals, four brigadiers and a raft of colonels to do its job. It probably needs one lieutenant general as a commander and a colonel as its chief of staff. When deployed, it might need one other senior officer. Some of the officers in the ranks of lieutenant colonel to major general spend much of their effort trying to make bloated HQs, full of poorly-trained staff, work.
The weakness of HQs does not appear to be a big problem at the moment. NATO, and the coalitions which its members tend to dominate, generally holds a monopoly in the use of large-scale collective violence. The fact that their HQs are poorly-trained, bloated and inefficient does have negative consequences, but they are rarely exposed. However, it will not always be that way. Some nation somewhere will be capable of doing simple but violent things quickly and effectively. When that happens, an army of a developed nation (or a coalition of such) will take a bad beating. But it doesn’t have to be that way.
The key lies in the people. There should be fewer of them, but they should be better trained and better organised. IT should help bind talents together as a team. Reduced to its absolute basics, the functions of an HQ (even the largest) are simple and can be delivered by a few people. They need to be the right people and be highly trained, both individually and as a team.
There should be a fundamental review of what HQs are required to do. It should start from the premise that staff only exist to assist a commander exercise command. If they don’t do that, they shouldn’t be there. The principle functions can be seen as leadership, decision making and control. What does that mean? How can that be delivered with the smallest practical number of people? Brook’s Law, or something like it, will not go away. There should then be a fundamental review of how digital IT can, and should, assist in the process.
CPs should focus on output leading to outcome. That means that within a theatre there is one, and only one, enduring campaign plan. Anything below that should be contained in short, succinct orders which describe simple, robust tactical plans.
Overall, there are two overarching requirements. The first is to understand that command is a fundamentally human issue. It is best delivered by small, expert groups of talents working together in teams. The second is to then accept that, at present, something is badly wrong.